At Productcaster, we have helped some of Europe’s largest retailers unlock this potential to either steal incremental sales from their competitors or improve revenue efficiencies through two clear marketing strategies to retain the 20% saving:
Strategy A – Reduce cost and maintain sales revenue (improved COS)
Strategy B – Maintain average CPC and increase revenue (no change in COS)
But what happens if you move from your CSS partner back to Google?
We have been proactively working with a small selection of our clients who wanted to test this and understand the true effects of CSS now that the Google ad credit has ended. We have a dedicated team of data scientists who work closely with our partner development team to help our clients understand the CPC benefits through our bespoke measurement framework. This allows us to launch new clients seamlessly and help them understand the true performance benefits of Productcaster.
What were the results?
One of our clients saw a 29% drop in impressions and traffic through Google for the same average CPC as they were achieving through Productcaster, with no changes to budget and bid optimisation following the switch to Google. As a result of this decline in visibility, the client saw a 11% drop in sales and a 9% decline in revenue in the two weeks immediately after the switch.
*Test methodology: Data taken from a 25-day period, normalised and indexed against the average of the pre-test behaviour.
-29% drop in impressions and traffic
-11% decline in sales
-9% decrease in revenue
For all our clients who tested using this approach, all have continued to use Productcaster as part of their Shopping strategy and recognised the importance of having a partner to help understand these results in more detail.
What’s the right measurement strategy?
Although the Google ad credit expired on December 31st, 2018, the CPC reductions afforded to Productcaster will remain in effect indefinitely. To understand the benefits of CSS, it’s important to have the correct measurement framework in place. We recommend:
- Running tests for a minimum 5-day period (depending on impressions volume) to gather enough data for accurate and meaningful results
- Normalising key metrics and analysing the data across pre and post launch periods
- Not running tests during periods of volatile spend or seasonal changes